Oodl.ai vs Learning Transformer

Oodl is a personal museum in your pocket — the effortless swipe of social media, but every page is a masterpiece. THE FEED, REIMAGINED Swipe up for the next work. Swipe sideways to change wings. Over a hundred thousand open-access artworks stream live from The Metropolitan Museum of Art, the Art Institute of Chicago, the Cleveland Museum of Art, and SMK — the National Gallery of Denmark. Full screen, no ads, no comments, no account. nOODL WHAT MOVES YOU Double-tap to nOodl (save) a work. Your saves build a private collection, teach the on-device blend your taste — more of the eras, mediums, and artists you respond to — and grow into a living map of your eye: a floating bubble portrait of your taste, stirred by the tilt of your phone. YOUR PHOTOS, ON THE SAME WALL Create private galleries from your photos and videos, or subscribe to a Photos album — including iCloud Shared Albums — and it becomes a channel that updates automatically as you add photos on any device. Videos loop full-screen like reels. Everything flows into your All feed alongside the museums. Oodl filters screenshots and near-duplicates on-device and surfaces your best shots. Double-tap your own photo to give it a title. LEARN THE STORY Tap "learn" on any piece for a personal art-magazine page: the subjects in the image (identified entirely on your iPhone), encyclopedia context on the artist, the work, and the place a photo was taken, and — on iPhones with Apple Intelligence — a short curator's note written on-device. In focus mode, pinch into the brushwork and let Apple's Visual Look Up identify what you're seeing. RIGHTS DONE RIGHT Every artwork carries its license (CC0 / Public Domain), museum credit line, and a link back to the source. Your own media is marked private and never leaves your device. FREE MEANS FREE No subscription, no ads, no tracking, no account. The world's museums opened their collections to everyone — Oodl is their home on your phone.

A Transformer, built by hand and shown one step at a time. Twenty-three screens follow "Attention Is All You Need" (Vaswani et al., 2017) in the paper's own order: a word becoming numbers, the three vectors, the scores, the softmax, four heads merged into one, the mask, the second attention, and the loop that writes one word at a time. What this is not: a trained model. Every weight here was set by hand, so that each one can be read in a single line. Nine numbers stand for a word instead of 512. One encoder layer and one decoder layer instead of six of each. Four heads, four words of vocabulary, one short sentence. What it is: the paper's arithmetic at a size you can check by hand. Every number on every screen is computed while you look at it — no videos, no quizzes, no account, no network. Steps 1 to 10 are free. Steps 11 to 23 are one purchase, made once and kept.